The risk of incident atherosclerotic cardiovascular disease (ASCVD) among patients with end-stage renal disease (ESRD) is between 5 and 20 times that seen in the general population. Many of the traditional Framingham risk factors, however, do not have the same risk profiles in ESRD as in the general population. Large studies of lipids and mortality in ESRD have found an inverse relationship to that seen in the general population, such that higher mortality is associated with lower total cholesterol (TC), low density lipoprotein cholesterol (LDL-C) and triglycerides (TG). The etiology of this inverse relationship is poorly understood, but in part results from the high degree of comorbid illness and malnutrition seen in the dialysis population. Overall mortality is associated with low body mass index (BMI), low albumin and low creatinine in ESRD. The relationship between lipids and incident ASCVD in ESRD may also be clouded by these same factors. Few prospective studies investigating the association between lipids [including TC,LDL-C, TG, high-density lipoprotein cholesterol (HDL-C), apolipoprotein-A1 (apo-A1) and apolipoprotein-B (apo-B)] and clinical ASCVD events in ESRD have been performed. Most of these studies recruited prevalent ESRD patients and adjusted for limited confounding variables. The CHOICE Study is comprised of 876 incident ESRD patients with a wide range of data on confounding variables in addition to serum banked within 3 months of enrollment into the study. This grant seeks funding for a lipid panel including TC, HDL-C, apo-A1, LDL-C, apo-B, and TG for 876 CHOICE participants to test the hypothesis that: Low HDL-C and apo-A1 levels, and high TC, LDL-C, apo-B and TG levels will predict incident ASCVD events, after adjustment for comorbidity, BMI, dialysis dose, creatinine and albumin. The rationale for this hypothesis is that if adjustment is made for these and other confounding factors, an underlying positive relationship between lipids and ASCVD may be observable.